Instructions to use ViorikaAI-org/MicroLlama with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use ViorikaAI-org/MicroLlama with Transformers:

# Use a pipeline as a high-level helper
from transformers import pipeline

pipe = pipeline("text-generation", model="ViorikaAI-org/MicroLlama")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("ViorikaAI-org/MicroLlama")
model = AutoModelForCausalLM.from_pretrained("ViorikaAI-org/MicroLlama")

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use ViorikaAI-org/MicroLlama with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "ViorikaAI-org/MicroLlama"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "ViorikaAI-org/MicroLlama",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/ViorikaAI-org/MicroLlama

SGLang

How to use ViorikaAI-org/MicroLlama with SGLang:

Install from pip and serve model

# Install SGLang from pip:
pip install sglang
# Start the SGLang server:
python3 -m sglang.launch_server \
    --model-path "ViorikaAI-org/MicroLlama" \
    --host 0.0.0.0 \
    --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "ViorikaAI-org/MicroLlama",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker images

docker run --gpus all \
    --shm-size 32g \
    -p 30000:30000 \
    -v ~/.cache/huggingface:/root/.cache/huggingface \
    --env "HF_TOKEN=<secret>" \
    --ipc=host \
    lmsysorg/sglang:latest \
    python3 -m sglang.launch_server \
        --model-path "ViorikaAI-org/MicroLlama" \
        --host 0.0.0.0 \
        --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "ViorikaAI-org/MicroLlama",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use ViorikaAI-org/MicroLlama with Docker Model Runner:
```
docker model run hf.co/ViorikaAI-org/MicroLlama
```

🦙 MicroLlama

Эксперементальная микро-модель

Подробнее:

⚙️ Детали модели

Архитектура: LLAMA
Параметры: 3M
Язык: Русский
Лицения: MIT

🏋️ Детали Тренировки

Датасет: ``
Железо: ОДНА NVIDIA GEFORCE RTX 5060 TI (16GB VRAM)
Эпохи: - 18
СРЕДНИЙ LOSS: 0.4349
Оптимизатор: 5e-4
Контекст: 32 токена

🏋️ Использование

from transformers import AutoModelForCausalLM, PreTrainedTokenizerFast

model_name = "ViorikaAI-org/MicroLlama"
tokenizer = PreTrainedTokenizerFast.from_pretrained(model_name)
model = AutoModelForCausalLM.from_pretrained(model_name)

model.eval()

prompt = "<s> Привет, как дела? =>"
inputs = tokenizer(prompt, return_tensors="pt")

outputs = model.generate(
    **inputs,
    max_new_tokens=32,
    temperature=0.7,
    top_k=50,
    top_p=0.9,
    do_sample=True,
    no_repeat_ngram_size=2,
    pad_token_id=tokenizer.pad_token_id,
    eos_token_id=tokenizer.eos_token_id
)

full_text = tokenizer.decode(outputs[0], skip_special_tokens=False)
answer = full_text.split("=>")[-1].replace("</s>", "").strip()

print(f"Ответ модели: {answer}")

🛜 Наши Соц. Сети

Discord: https://discord.gg/8JwTv8zj8d , https://discord.gg/7JE7maH6cf
Telegram: https://t.me/viorika_official

Downloads last month: 27

Safetensors

Model size

3.65M params

Tensor type

F32